metadata

license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_3x_deit_tiny_adamax_001_fold3
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: test
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.905

smids_3x_deit_tiny_adamax_001_fold3

This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.9948
Accuracy: 0.905

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.001
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.542	1.0	225	0.4548	0.81
0.3403	2.0	450	0.3948	0.8633
0.3018	3.0	675	0.3258	0.88
0.2181	4.0	900	0.3725	0.8583
0.2784	5.0	1125	0.3487	0.8667
0.2253	6.0	1350	0.3694	0.87
0.1182	7.0	1575	0.4281	0.8683
0.1479	8.0	1800	0.4669	0.8683
0.127	9.0	2025	0.3858	0.88
0.1437	10.0	2250	0.6727	0.825
0.1318	11.0	2475	0.5423	0.8583
0.1039	12.0	2700	0.5755	0.8717
0.0315	13.0	2925	0.6762	0.8633
0.0565	14.0	3150	0.6056	0.8833
0.0169	15.0	3375	0.6739	0.8667
0.0394	16.0	3600	0.7747	0.87
0.051	17.0	3825	0.7121	0.8817
0.0214	18.0	4050	0.7547	0.88
0.0367	19.0	4275	0.7020	0.8583
0.0574	20.0	4500	0.7090	0.8783
0.016	21.0	4725	0.8561	0.87
0.0011	22.0	4950	0.6767	0.8783
0.0009	23.0	5175	0.6981	0.89
0.0024	24.0	5400	0.8528	0.8717
0.0185	25.0	5625	0.7739	0.8833
0.0018	26.0	5850	0.9050	0.875
0.0011	27.0	6075	0.8197	0.8767
0.0199	28.0	6300	0.8264	0.8833
0.0076	29.0	6525	0.8894	0.895
0.0073	30.0	6750	0.8362	0.9
0.004	31.0	6975	0.8565	0.9033
0.0	32.0	7200	0.9512	0.8967
0.0	33.0	7425	0.8488	0.895
0.0	34.0	7650	0.8884	0.9033
0.0	35.0	7875	1.0628	0.8917
0.0	36.0	8100	0.8726	0.9017
0.0029	37.0	8325	0.9056	0.9067
0.0	38.0	8550	0.9531	0.9033
0.0	39.0	8775	0.9541	0.905
0.0	40.0	9000	0.9488	0.905
0.0	41.0	9225	0.9370	0.9083
0.0	42.0	9450	0.9567	0.9067
0.0	43.0	9675	0.9765	0.9033
0.0	44.0	9900	0.9911	0.9017
0.0	45.0	10125	0.9807	0.905
0.0	46.0	10350	0.9732	0.9083
0.0026	47.0	10575	0.9856	0.905
0.0	48.0	10800	0.9870	0.905
0.0	49.0	11025	0.9903	0.905
0.0	50.0	11250	0.9948	0.905

Framework versions

Transformers 4.32.1
Pytorch 2.1.1+cu121
Datasets 2.12.0
Tokenizers 0.13.2